Talk:History of the Romanian language

"traits of many changes of the Latin which occurred in the 4th and 6th centuries"

I've requested examples for this claim more than a few months ago. It does not seem to appear anywhere in standard academic literature at the moment, in fact the understanding now seems to point at the opposite:

"Developments appear to have been significantly different in the Balkans, in particular in the sub-Danubian area, where the Daco-Romance varieties can be reasonably assumed to have arisen." - Barbato - The Early History of Romance Palatalization

As such the claim attributed to Vekony might be a minority view and should not be the lead of the paragraph.

Leaving this entry here in case other editors want to discuss about it. Aristeus01 (talk) 20:18, 13 June 2024 (UTC)

Unexplained palatalisation?

The Palatalisation section under Internal History gives the examples deosum > *di̯ósu > *djosu > jos and hordeum > *órdi̯u > ordzu > orz, which seem to show two different reflexes of /dj/; Why does it become /ʒ/ (presumably through intermediary /d͡ʒ/) in some words, but /d͡z/ → /z/ in others? I see that an earlier version of the article answers this question (albeit without a source), however this explanation was later removed, being replaced with information from another source, "The Early History of Romance Palatalizations" by Marcello Barbato, I cannot access this source (which does not seem to be cited on the page itself), so I do not know if it also answers this question, but if it does I request that that explanation be added to the article (and this source be cited in it).

Alternatively, is the word "Jos" perhaps instead derived from another form of the Latin word, "iosum", as Wiktionary suggests, with the /d/ already dropped, making Romanian initial /ʒ/ a derivative solely of Latin /j/ rather than /deV diV/? (Later reinforced by borrowings from other languages; All other Romanian 'j'-initial words of Latin origin I can find on Wiktionary are derived from Latin /j/) If this is the case, the page should likely be edited to remove that as an example of palatalisation of /d/.

Listed on another page however, Palatalization in the Romance languages, I see a claim (With a different source: "Palatalizations in the Romance Languages" by Daniel Recasens) that both /dj/ and and /j/ became /ʒ/ in initial position during the transition from Latin to Romanian, but, when intervocalic, /dj/ became /z/ instead, which would also explain the discrepancy. As I have access to neither of the sources I've mentioned, and additionally cannot find sources for two of these claims, I cannot properly assess the different theories to determine which is most likely.

EDIT: Upon further consideration, Recasens' explanation (The only one I can actually find a source for, assuming his work does indeed explain this) doesn't add up, due to the common word "Zi" /zi/, from Latin "Dies", which we'd expect to instead be /ʒi/ if this was the case, unless the original /i/ was retained rather than becoming /j/, and this is in fact a result of secondary palatalisation. Additionally I've found a few other examples of initial /dj/ → /z/, as well as the word "Zăcea", which seems to display Latin /j/ becoming /z/ word-initially, rather than remaining as /j/ like Palatalization in the Romance languages claims, or becoming /ʒ/ as I theorised before. Since "Jos" is the only example I have found of /dj/ → /ʒ/, it so far seems most likely that theory 2 is correct, however this does not explain the irregular change of /j/ → /z/ found in "Zăcea".

I also see, in another section of the page, the diphthong ⟨âi⟩ /ɨj/ is proposed as having potentially originated due to palatalisation followed by metathesis, but under what context would there be palatalisation here? Does /n/ regularly palatalise before /e/? Or is it caused by the specific environment of /ɨne/? If so, are there other examples of this, including any without metathesis? Or do all dialects where it's metathesised always do such?

The latter seems most likely due to the dialectal shift ține > țâni̯e, which shows palatalisation of /n/ (Or breaking of /e/) in combination with the preceding /i/ being centralised, although if this is the case I feel it should be mentioned in the text of the page, rather than vaguely implied. Erated8 (talk) 06:08, 21 October 2024 (UTC)

Latin "dies" (accuative "diem") is not relevant to the evolution of /dj/, because it did indeed retain /i/ rather than evolving into /djes, djem/. In Romanian /di/ palatalized to /zi/, as in Latin dicere > Romanian a zice, but /d/ was retained here in other Romance languages: compare French vendredi and Italian venerdì from Latin Veneris diem. The /i/ seems to have been retained as a syllabic vowel in "dies" and "diem" because the first syllable of these forms was stressed: /ˈdi.eːs/, /ˈdi.em/. In contrast, deosum was stressed on the second syllable.--Urszag (talk) 17:51, 23 October 2024 (UTC)